Jump to content

Talk:Comparison of H.264 and VC-1

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

Goals section in table modified

[edit]
  1. I modified the Goals section, since it was not cited anyway. The previous goal for H.264 made it look as if it was designed to be an inferior codec to VC-1, as it allowed flexibility but not necessarily quality. The rewritten goal, while relatively long (I could not find a way to encase it in a sentence), makes clear that the different profiles do allow for flexibility (such as usage on portable devices), but the high profiles aim at high (efficient) compression and quality, which suggests, standard versus standard, it could rival VC-1 at the high levels. 77.126.249.55 (talk) 17:32, 30 August 2008 (UTC)[reply]

Additional references

[edit]
  1. Templates requesting additional references should not be removed until AFTER a change is made or a compromise is reached - not before. Two different editors have noticed the same thing, so there is definitely the perception that there may a problem here.
  2. The biggest problem that I have, and I believe the other editor had, was that all the subjective items on the page are sourced by microsoft.com, the developer of VC-1. A strong attempt needs to be made to find other sources for these subjective evaluations.
  3. Citing the number of figures in a specification has almost no meaning on the quality or understandability of the documentation. This, and other things, make this article boarder on original research, which is not allowed in wikipedia. Respected sources need to be found that have already made this assertion, and those sources should be cited.
  4. What are the "Performance demands" for H.264? What does "more PC's" mean? A specific example should be mentioned and cited. As it stands right now, the reader is left with either the assumption that the writer only knows about VC-1 (in which case, how could a valid comparison be made?), or information about H.264 was deliberately left out. You and I know that H.264 has higher complexity - but how much higher? Will a E4300 C2D do 1080p with VC-1 but not H.264?
  5. Also, bit rate is a valid comparison - please leave it in the table. If you don't agree with which one is better, either request a cite, or change it with a counter citation. My understanding is that H.264's higher complexity results in a lower bit rate.

Thank you! Mrand T-C 04:09, 15 June 2007 (UTC)[reply]

Reference discussion

[edit]
  1. references should not be removed until AFTER a change is made
    The page was edited each time after the such comment was made.
    Don't get me wrong, I'm very grateful for the article and the time that you've spent on it, but saying that the page was edited each time is like following the letter, rather than the intent, of the template request. In both cases, I think the concern was clearly stated about the overwhelming use of only one source for subjective comparisons (see below), and that one source happens to be the developer of the technology. Since the concern remains, the template needs to remain as well.
  2. that all the subjective items on the page are sourced by microsoft.com
    I don't know what you are calling subjective. Test results are good reference material. Also if someone has a better product they are going to advertise it. If someone has a worse product they are not going to advertise that it is worse.
    Absolutely test results are good material. But I suspect that we are both well educated professionals here, so we both know that "video quality" can be a very subjective test and that there are many different way to configure the encoder, and therefore many way to make one encoder come out on top of the other. The other editor, and myself, are simply requesting additional sources to back up the quality claims.
  3. those sources should be cited (context: with regard to readability of the specifications)
    They are cited. How better do you want them cited? There really is no comparison. Anyone who has looked at both no what a joke the h.264 spec is compared to VC-1.
    Unfortunately, counting figures is not considered a valid way to make the claim that one is more readable than the other. The H.264 spec may very well be completely unreadable to even the most technical person, but we (editors) are not allowed to make that determination. Our job is to find and capture reliable information from reliable sources. If wikipedia allowed editors to make judgements like this, it would not be NPOV.
    Comments are out of date. All the current text does is count number of figures. Daniel.Cardenas 16:58, 16 June 2007 (UTC)[reply]
  4. What are the "Performance demands" for H.264? What does "more PC's" mean? A specific example should be mentioned and cited.
    Yes, but what is stated is cited. I could state that on several of my computers VC-1 displays well and h.264 1080p stutters but that wouldn't be satisfactory. I believe it is obvious to anyone who has done minimal testing. I believe this has more to do with how well the decoder is optimized rather than the compression standard itself.
    I'm sorry - I do not see anywhere on the microsoft pages which state that VC-1 is usable on "more PC's". Could you point out the particular paragraph? I do see where they say "today's off-the-shelf computer hardware", but that is still an almost useless as a piece of information from the perspective of a world-wide encyclopedia that will be read for years to come. What is "today's computer?" A Mac? A 1GHz laptop? A 3 GHz Desktop? Thankfully, PC changes over time - but since they do, it makes our job somewhat more difficult because we have to find a source (and preferrably multiple) which describe the type of hardware typically required to display 1080p. I also try to think of it from the readers point of view (but not an overly technical reader): Wikipedia says a modern PC can display 1080p. The reader just bought a computer last year - so they believe it's modern. But will it play the content? You and I know that it depends on how the content is encoded, the exact decoder being used, and the computer being used.
    Have you conducted your own tests? If you have you will know how true it is. There is also a subjective aspect to this, how many drop frames is o.k.? Daniel.Cardenas 16:58, 16 June 2007 (UTC)[reply]
  5. Also, bit rate is a valid comparison - please leave it in the table. If you don't agree with which one is better, either request a cite, or
    Its uncited text and per wikipedia policy anyone can delete it. It also happens to be wrong. The references from the microsoft website state that at same bit rate VC-1 is judged to be of better quality. Would it be productive that I add another line stating the opposite with a reference while your line doesn't have a reference and never will? Wikipedia:Citing_sources#Unsourced_material - Do not be inappropriately cautious about removing unsourced material
    Could you please point out what paragraph the microsoft site (or any other reference) which states that this claim is incorrect? I see on the microsoft site that comparisons are made to MPEG-2 bitrate, and few to MPEG-4 (which I believe refers to MPEG-4 part 2, but very few to H.264 - and those that do reference the video quality of H.264 baseline with VC-1 mainline and make no mention of bit rate.
    Standard operating procedure when conducting these tests is to use the same bitrate. Daniel.Cardenas 16:58, 16 June 2007 (UTC)[reply]
    VC-1 reference spec came 3 years after h.264. Do you think in the fast pace world of computer science they wouldn't make a better product in this time? It would be like expected mpeg2 to be better than h.264. What do you think the reason is that both blu-ray and hd-dvd movies use VC-1 and not h.264? Daniel.Cardenas 06:04, 15 June 2007 (UTC)[reply]
    Unfortunately, it doesn't matter what you or I think (if only it did, it would make this job so much easier... except that more editor warring would go on throughout wikipedia)! Anything published here can not be original research - it must be verifiable from elsewhere. I have also never read that no blu-ray or HD-DVD movies are yet encoded with H.264. That would be a very good mention for this article if you know of a source! Mrand T-C 14:13, 16 June 2007 (UTC)[reply]
    Oops, I found out I was wrong about h.264 and blu-ray. My apologizes. hd-dvd is about 85% VC-1 but blu-ray is about 85% h.264, 10% mpeg2. (embarrased). I mis-read a work email that was sent out. Daniel.Cardenas 19:27, 19 June 2007 (UTC)[reply]
    That ratio doesn't sound correct. Many BluRay titles use VC-1 because the studios either re-used their original HD-DVD encodings (the bitstreams are still valid for BluRay) or because they were more familiar with the VC-1 encoding software. The number of VC-1 encoded Blu-Ray titles is much higher than 5 percent. —Preceding unsigned comment added by 131.107.0.81 (talk) 23:55, 28 January 2011 (UTC)[reply]

Subjectivity

[edit]

At worst, this article smacks of POV and at best it is original research. The article cites too many subjective tests. I am critical of the overwhelming Microsoft bias, but I don't think that this should be corrected by inclusion of findings of other groups - the compilation of such work is original research. I think that the only way to avoid this subjectivity is to produce a comparision of the published specification for each codec and remove the subjective comparisons. Also, the inclusion of design goals and such is redundant and should be removed. None of these factors indicate the actual functioning of either codec and in this context seem only to present POV.

Please try to persuade me otherwise. Senordingdong 16:33, 16 June 2007 (UTC)[reply]

Which subjective comparisons are you referring to? Which POV? Which original research? I'm not seeing what you are seeing.
Here is independent conformation from the the Microsoft site:
C'T Magazine, Germany's premiere audio-video magazine, compared various codecs, including VC-1, H.264, and MPEG-2, and selected VC-1 as producing the best subjective and objective quality for high-definition (HD) video.

The article cites too many subjective tests.

How many is too many? Codec comparisons are typically done subjectively. Standard industry practice.

the compilation of such work is original research

Your definition of original research doesn't match Wikipedia's.

the inclusion of design goals and such is redundant and should be removed.

Its a good summary that helps people understand were these codecs are coming from and what maybe there best use.
Daniel.Cardenas 17:05, 16 June 2007 (UTC)[reply]

RE: subjectivity

[edit]

The article cites too many subjective tests.

How many is too many? Codec comparisons are typically done subjectively. Standard industry practice.
  • Any is too many. It may be standard practice, but they are all highly subjective and are thus unsuitable for inclusion within an encyclopedia (POV).

the compilation of such work is original research

Your definition of original research doesn't match Wikipedia's.

the inclusion of design goals and such is redundant and should be removed.

Its a good summary that helps people understand were these codecs are coming from and what maybe there best use.
  • It gives no information about the actual performance or suitability of the current codec for a given task. This can only be determined by considering the codec in its final and delivered state.

Senordingdong 17:54, 16 June 2007 (UTC)[reply]

Which synthesis are you referring to?

[edit]

You asked, 'Which synthesis are you referring to?' I said that, 'I am critical of the overwhelming Microsoft bias, but I don't think that this should be corrected by inclusion of findings of other groups - the compilation of such work is original research'. This is the synthesis that I refer to. The compilation/interpretation of such results is considered original research under WP:NOR under the heading 'Synthesis of published material serving to advance a position'. Accordingly, it should not be included. My point is that including the findings of other groups in the article would be a poor correction to the Microsoft bias of the page.

Senordingdong 13:56, 17 June 2007 (UTC)[reply]

Thanks for the clarification. I believe others will find that your viewpoint is unique, and is not what is intended by wikipedia policy. In otherwords there is no synthesis. You seem to be claiming because the information is from one source that is synthesis, and that is not is not what is meant in the policy. And then you further try to suggest that if the same data is from two sources then it is still not valid. Doesn't make any sense to me. Daniel.Cardenas 17:09, 17 June 2007 (UTC)[reply]
I'm not claiming than any synthesis exits. I state that there is Microsoft bias on the page and that this bias should be removed. One way to remove this bias would be to include the findings of other groups along with the current Microsoft citations. However, interpreting or combining such results would be a synthesis of published material, and would violate WP:NOR. Therefore this cannot be done. I propose that a better solution is to remove the subjective tests and citations and instead compare codecs based purely on their specifications. In this way, we remove subjectivity and do not include POV in the article. I do not believe that this viewpoint is unique nor that it contradicts the spirit of Wikipedia policy. Senordingdong 18:28, 17 June 2007 (UTC)[reply]
Group results based on perception or subjectivity is a standard scientific and engineering practice. I'll add the reference from C'T magazine; that will address your request for adding findings from other groups. Are you claiming an incorrect bias for VC-1 video quality? If yes, then find a reference, subjective or objective that says so. Daniel.Cardenas 18:37, 17 June 2007 (UTC)[reply]

just in case you're interested in - http://compression.ru/video/codec_comparison/subjective_codecs_comparison_en.html

This article needs a serious rewrite

[edit]

This could end up being a very useful article, but right now it seems put together very haphazardly. I would suggest:

1. Moving ALL references to the very end of the article because currently they make the article very hard to read.

2. Separate objective specification differences from subjective design goal and application differences. Add another chapter which discusses the suggested advantages of each codec's specification differences, complete with references.

3. Add a final chapter which discusses current implementation differences (real-world differences, not design/spec differences).


1. Moving ALL references to the very end of the article because currently they make the article very hard to read.
That was a boo boo of mine. I deleted it. Thanks for pointing that out. Although probably changing all references to use the ref tag is needed. Daniel.Cardenas 01:48, 13 July 2007 (UTC)[reply]

This article has some serious issues

[edit]

Most of the information is horribly out of date. A reference stating that VC-1 is better than H.264 (or vice versa) is completely useless if it comes from 2003, before there were any worthwhile encoders for either. Its like using a reference from 1950 to make the claim that human space travel is impossible. Is there consensus to do a wipe of all the references that are too old to be valid for modern comparisons? —Dark•Shikari[T] 09:39, 12 November 2007 (UTC)[reply]

If you wipe the reference, then you will have to wipe the associated comments. Ideally it would be good if there was an alternative reference available, even if it changes the comparison. Go for it. Thanks, Daniel.Cardenas 12:20, 12 November 2007 (UTC)[reply]
Done. Also note that it is probably a violation of NPOV policy to use any source by a party with a strong vested interest (such as Microsoft, Mainconcept, Elecard, etc) for anything controversial. For example, if Microsoft makes the claim that VC-1 is faster than H.264, this is not necessarily a fact; I've seen benchmarks demonstrating the opposite with CoreAVC used as the H.264 decoder. On the other hand, one cannot trust a page by Mainconcept written about how good Mainconcept is, either. Such sources are obviously by parties with a vested interest and as such can only be used for purely unbiased facts, such as technical specifications. —Dark•Shikari[T] 17:11, 12 November 2007 (UTC)[reply]

Possible to add real world uses

[edit]

Perhaps not fully needed, as it may change over time, but if there were a section that briefly states where one can usually find each codec being used (i.e, industry, personal backups, HD-DVD, etc) it could prove to be useful. —Preceding unsigned comment added by 67.71.252.98 (talk) 18:06, 9 September 2008 (UTC)[reply]

[edit]

Hello fellow Wikipedians,

I have just modified 2 external links on Comparison of H.264 and VC-1. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}} (last update: 5 June 2024).

  • If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
  • If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—InternetArchiveBot (Report bug) 16:23, 11 August 2017 (UTC)[reply]